Nailing down PKC isoform specificity in diabetic nephropathy two's company, three's a crowd

Diabetic nephropathy is characterized by early vascular dysfunction and increasing matrix accumulation in the kidney, eventually leading to proteinuria, glomerulosclerosis and interstitial fibrosis [ 1 ]. Over the last decade, our understanding of the molecular mechanisms and the pathogenesis of diabetic nephropathy (and other diabetic microvascular complications) has been greatly enhanced [ 2 ]. However, we still do not completely understand how metabolic disturbances in the diabetic state, i.e. hyperglycaemia, induce such a vast array of distinct cellular events leading to progressive renal failure [ 2 ]. Several hypotheses linking hyperglycaemia and altered cellular biology have been proposed [ 3 ]. One of these hypotheses postulates that high glucose concentration leads to the activation of the calcium- and phospholipid-dependent protein kinase C (PKC) signalling pathway which subsequently mediates cellular response, e.g. with altered gene expression [ 4 ]. It is generally believed that intracellular PKC activation is achieved by the diabetes-induced accumulation of one of its co-factors, diacylglycerol (DAG), inside the cell [ 2 , 5 ]. Diabetes mellitus causes elevated DAG levels in vascular tissues associated with diabetic complications, including retina, heart, aorta and renal glomeruli, and in non-vascular tissues such as liver and skeletal muscle [ 2 ]. However, PKC may also be activated by other mechanisms [ 6 , 7 ]. Oxidative stress has been reported to induce prolonged activation of PKC within cells [ 8 ] through reactive oxygen species (ROS), produced by hyperglycaemia or through ‘advanced glycation end products’ (AGEs) which have been shown to directly activate PKC [ 9 , 10 ]. Since PKC is an important intracellular messenger system and plays a central role in cell proliferation, matrix expression, apoptosis and regulation of gene transcription [ 4 ], PKC activation as one of the important underlying molecular mechanism of diabetic complications is an attractive hypothesis and a multitude of reports suggest such a mechanism [ 4 ]. Further support comes from experimental and human studies, where inhibition of PKC by specific drugs has been shown to prevent early changes in the diabetic retina and kidney [ 11–16 ]. However, PKC is not a single entity but consists of a family of at least 12 serine-threonine kinases with distinct co-factor activation, expression patterns and cellular functions [ 17 ]. These isoforms were first cloned in 1986 and have been divided on the basis of their regulatory domains into three larger subgroups: the classical (conventional) PKC isoforms α , β I/II and γ (regulated by calcium and DAG), the novel PKC isoforms δ , ε , η and θ (regulated by DAG) and the atypical (non-calcium-/non-DAG-regulated) PKC isoforms ζ and ι / λ ( Figure 1 ) [ 4 ]. Multiple studies over the last decade have clearly demonstrated that the various PKC isoforms have distinct cellular functions in different cell types, cellular compartments and signalling pathways [ 4 , 18 ]. Structure of PKC isoforms according to subgroups. The regulatory and catalytic domain are displayed. Four conserved (C1 to C4) and five variable (V1 and V5) regions are indicated. The isoforms of PKCβI and PKCβII are derived from a single gene by alternative splicing. Since specific PKC isoforms activated by hyperglycaemia vary across tissues and cell types [ 19 , 20 ] and several PKC isoforms could be involved in diabetes-induced organ damage, the identification of the individual PKC isoform(s) responsible for the long-term damage of hyperglycaemia has been a hotly debated issue over the last 20 years. In addition to its pathophysiological importance, this issue is also of great interest therapeutically, since specific PKC isoform inhibitors have been developed by several pharmaceutical companies [ 21 ]. The first culprit in the PKC isoform debate was the PKCβ isoform. Studies from King et al. [ 2 ] have proposed that inhibition of the two splice variants of the PKCβ gene (PKCβI and PKCβII) can prevent diabetic microvascular complications such as retino-, neuro- and nephropathy [ 11–13 , 19 , 20 ]. In experimental diabetic nephropathy, PKCβ inhibition has been related to the functional and histological abnormalities of glomeruli in two different diabetic animal models [ 12 , 13 , 19 , 20 ]. These results are largely based on in vivo studies, using a pharmacological inhibitor which is suggested to be selective for the PKCβI- and -βII-isoforms (Ruboxistaurin®, LY333531) [ 22 ]. In the first study, oral feeding of this compound to streptozotocin (STZ)-induced diabetic rats, a rodent model of Type 1 diabetes mellitus, prevented the increased glomerular mRNA expression of transforming growth factor-β1 (TGF-β1) and extracellular matrix components such as fibronectin and collagen IV [ 12 ]. In a second study, oral administration of Ruboxistaurin® in db/db mice, a mouse model of Type 2 diabetes mellitus, inhibited mesangial expansion [ 13 ]. In another recent study, in vivo inhibition of PKCβ with Ruboxistaurin® in a transgenic rat model (Ren-2) led to a reduction in albuminuria and TGF-β1 expression, as well as less structural injury despite continued hypertension and hyperglycaemia [ 14 ]. However, despite the impressive arguments for the PKCβ isoform in the pathogenesis of renal complications in diabetes mellitus, several investigators had other champions. Whiteside et al. [ 19 ] suggested that also high-glucose-induced activation of novel and atypical PKCs may contribute to early changes in diabetic nephropathy in rats. Immunoblotting of mesangial cellular fractions revealed increased membrane and nuclear expression levels of the PKC isoforms α, δ, and ε after 48 h in 30 mM glucose [ 23 ] and enhanced recovery of membrane-associated PKCζ as early as 24 h of high-glucose exposure [ 24 ]. Furthermore, it has been shown that mesangial cell filamentous actin disassembly and hypocontractility in high glucose are mediated by PKCζ [ 24 ]. Ha et al. [ 25 ] largely confirmed these results while demonstrating that the PKC isoforms δ and ε delta are sensitively activated by hyperglycaemia-induced oxidative stress in diabetic rat kidney. In 1999, Kang et al. [ 26 ] demonstrated that the PKC isoforms α and ε are activated in kidneys from STZ-induced diabetic rats, but failed to prove increased expression or activation of the two PKCβ splice variants. The case for the PKCα isoform was previously opened by Kikkawa et al. [ 27 ] demonstrating that activation of PKCα and ζ does occur in rat glomerular mesangial cultured under high glucose condition. Hempel et al. [ 28 ] have shown that an increase in extracellular glucose leads to a rapid dose-dependent increase in endothelial cell permeability via a PKCα-dependent signalling pathway. A functional role of PKCα was further supported by in vitro experiments demonstrating that high glucose leads to an PKCα-dependent increase of TGFβ1 expression in cultured vascular smooth muscle cells [ 29 ]. Since PKC alpha is also activated by oxygen free radicals [ 30 ], this isoform also seemed to be a good candidate for the link between hyperglycaemia and cellular activation in diabetes mellitus. Several years ago, a new avenue was opened for researchers to solve the problem of PKC isoform specificity in diabetes-induced organ damage, when Leitges et al. [ 31 ] generated PKC-isoform-specific knock-out (KO) mice. Using these animal models, a specific role of individual PKC isoforms such as PKCα and β in the development of diabetic nephropathy was identified and a novel hypothesis on the pathogenesis was generated ( Figure 2 ). First, it was shown that the development of albuminuria in STZ-induced murine diabetic nephropathy is regulated via a PKCα-dependent signalling event [ 32 ]. We demonstrated that hyperglycaemia in PKCα KO mice does not lead to increased of the permeability of the glomerular filtration barrier, comprised by the endothelial cell layer, the glomerular basement membrane and the glomerular visceral epithelial cell (podocyte) layer [ 32 ]. High-glucose-induced activation of the PKCα isoform thereby seems to mediate its effect on the development of albuminuria via three different molecular mechanisms: (i) First, the high-glucose-induced increase of VEGF and its receptors is not present in PKCα null mice [ 32 ]. Since VEGF has been implicated in the pathogenesis of diabetic nephropathy, this mechanism may contribute to the PKCα mediated renal injury [ 33 ]. (ii) Secondly, the high-glucose-induced loss of the heparan sulphate proteoglycans such as perlecan in the glomerular basement membrane was completely inhibited in PKCα KO mice [ 32 ]. Since proteoglycans are the carriers of negative charges in the basal membrane, the persistence of these molecules under diabetic conditions in PKCα-deficient mice may also explain the lack of albuminuria [ 34 ]. (iii) Most interestingly, deletion of PKCα in vivo abolishes nephrin loss in STZ-induced murine diabetic nephropathy [ 35 ]. We have shown that this effect is due to a PKCα-dependent down-regulation of the transcription factor WT-1 in diabetic kidneys, a mechanism which is prevented when PKCα is blocked [ 35 ]. The PKCα isoform therefore seems to be an important and prominent mediator of glucose-induced glomerular changes. However, despite the striking findings regarding structural and functional changes in the glomerular filtration barrier, we also observed the increased renal and glomerular hypertrophy and the augmented TGF-β1 expression in the diabetic state to be unaltered in diabetic PKCα KO mice [ 32 ]. These observations are also confirmed by the findings of Ziyadeh et al. [ 36 ] who demonstrated that short- and long-term inhibition of TGF-β by neutralizing antibodies or treatment with antisense oligonuleotides prevents the development of renal hypertrophy but not albuminuria. Further studies pointed out that a comparable blockade of VEGF with systemic antibody administration inversely decreased albuminuria in STZ-induced diabetic rats or db/db mice with only moderate influence on mesangial expansion and renal hypertrophy [ 37 , 38 ]. Current concept of the molecular mechanism of diabetic nephropathy. Lately, we were able to demonstrate that this profibrotic pathway seems to be regulated by the PKCβ isoform, when we studied non-diabetic and STZ-induced diabetic PKCβ KO mice compared with appropriate 129/SV wild-type controls [ 39 ]. After 8 weeks of diabetes mellitus the high-glucose-induced renal and glomerular hypertrophy as well as the increased expression of extracellular matrix proteins such as collagen III and IV as well as fibronectin was reduced in PKCβ null mice [ 39 ]. Furthermore, the high-glucose-induced expression of the profibrotic cytokine TGF-β1 and CTGF were significantly diminished in the diabetic PKCβ KO mice in comparison with diabetic WT mice, suggesting a role of the PKCβ isoform in the regulation of renal hypertrophy [ 39 ]. Notably, increased urinary albumin/creatinine ratio persisted in the diabetic mice, even when the PKCβ gene was deleted in vivo [ 39 ]. Furthermore, loss of the basement membrane proteoglycan perlecan and the podocytary protein nephrin in the diabetic state was not prevented in the PKCβ null mice, as previously demonstrated in the non-albuminuric diabetic PKCα KO mice [ 32 , 35 , 39 ]. Recently, King et al. [ 39 ] presented a study from the Joslin Diabetes Center on renal pathophysiology in the STZ-induced diabetic mouse kidney, while also showing persistent albuminuria in the diabetic PKCβ KO mice and confirming our results. It has been suggested that deletion of the PKCβ isoform leads to improved renal dysfunction and pathology, while preventing an increased expression of NADPH oxidase complexes [ 40 ]. The differential effects of classical PKC isoforms regarding the development of diabetic nephropathy in our PKCα and PKCβ KO models suggest that activation of the PKCβ isoform contributes to high-glucose-induced, TGF-β1-mediated renal hypertrophy and fibrosis whereas perlecan as well as nephrin expression are regulated by a PKC-α-dependent signalling pathway, leading to diabetic albuminuria. These promising results regarding a specific role of both classical PKC isoforms in the development of diabetic nephropathy raised even more questions on the role of high-glucose-induced activation of novel and atypical PKC isoforms, as suggested by Whiteside and Dlugosz [ 19 ]. Surprisingly, analysis of the renal phenotype of PKCε KO mice with regard to renal hypertrophy and fibrosis revealed that the kidney/body weight ratio did not show any significant group difference compared with appropriate wild-type controls [ 41 ]. Urinary albumin/creatinine ratio remained normal in WT rodents, whereas PKCε null mice showed elevated albuminuria at ages 6 and 16 weeks [ 41 ]. Masson–Goldner staining revealed that tubulointerstitial fibrosis and mesangial expansion was significantly increased in the PKC-ε KO mice, even in the non-diabetic state [ 41 ]. However, this profibrotic phenotype has not been observed in other organs such as liver and lung [ 41 ]. Immunohistochemistry of the kidneys from PKC-ε null mice showed increased renal fibronectin and collagen IV expression which was further aggravated in the STZ-induced diabetic stress model [ 41 ]. Furthermore, TGF-β 1 expression and its activation measured as phospho-Smad2 and phospho-p38MAPK expression was increased in the PKC-ε KO mice, suggesting an inhibiting role of PKC-ε in TGF-β 1 and its signalling pathway in the (diabetic) kidney [ 41 ]. The latter results indicate that deletion but not activation of the PKC-ε isoform mediates renal fibrosis, possibly through interference with TGF-β1 signalling pathway. Therefore, activation of PKC-ε in the diabetic state may represent a protective response-to-injury, rather than being a mediator of renal injury. Derived from these latest in vivo studies, we postulate that hyperglycaemia leads to activation of the common intracellular PKC signalling pathway and its different isoforms ( Figure 3 ). PKC isoform specificity leads to diverge outcome leading to albuminuria and/or renal fibrosis in diabetic nephropathy. In conclusion, it has been demonstrated that hyperglycaemia activates several PKC isoforms, probably depending on the individual cell type or tissue, which subsequently interfere with progressive diabetic renal disease through independent mechanisms. PKC isoform specificity and cellular diversity seem to be responsible for the divergent outcome leading to albuminuria and/or renal fibrosis. Defining and affecting mediators of increased intracellular activation in diabetic myelopathy will help to design novel effective therapies in this common and complex chronic renal disease. The more is yet to come … Conflict of interest statement . None declared.